6 ms·

Chromatron Wi-Fi Pixel Controller

80 points by luu 7y ago

jkestner 7y ago
I have two of these. Way faster than an Arduino if you just want to set up some dynamic lighting, and still plenty of room for tinkering.
- lepouet 7y ago
  where did you buy it ?
  - jkestner 7y ago
    The Kickstarter. https://www.kickstarter.com/projects/1286098094/chromatron-wifi-pixel-controller https://www.kickstarter.com/projects/1286098094/chromatron-w...
paranoidrobot 7y ago
Did I just miss it, or is there no information about price, where to get it, whether it's available now, etc? It looks interesting, and if it were the right price I might actually want one. As far as I can tell there's only the signup for more info (no, thanks) - or some links to github and some social media.
- readmodifywrite 7y ago
  Hi, project creator here. Sadly, the poster was a little behind on catching this... The Kickstarter boards shipped over 2 years ago, and as the hardware didn't really sell well after that, it's mostly just a personal project that this point. That being said, I am actively working on a port for some stock ESP8266 boards (Adafruit Feather, NodeMCU, etc). I don't really want to sell hardware, but I'd love to see someone else using the software.
tootie 7y ago
I love that it has a terminal block. Secure connection and no soldering.
zeva 7y ago
seems like this may no longer be in production. https://kickstarter.com/projects/1286098094/chromatron-wifi-pixel-controller https://kickstarter.com/projects/1286098094/chromatron-wifi-... https://www.tindie.com/products/readmodifywrite/chromatron-wifi-led-pixel-controller/ https://www.tindie.com/products/readmodifywrite/chromatron-w...
- ShakataGaNai 7y ago
  Awww, that's very unfortunate. Found the repo ( https://github.com/sapphireos/chromatron https://github.com/sapphireos/chromatron ) since it's open source and there are no commits in the last 2 years.
  - readmodifywrite 7y ago
    Hi! I'm the project creator. While it's true the Kickstarter boards have all sold out, the project is still alive and well. Check out the dev and esp8266 branches, there's commits as of today ;-) My long term goal for this year is to get a port running on some stock ESP8266 boards (such as NodeMCU and the Adafruit Feather Huzzah). My original hardware design was too complex to build in a cost effective way, but I'd really like to get the software in the hands of those who are interested.
ShakataGaNai 7y ago
Very awesome demo. I want to do more DIY LED lighting in my office (because it makes for a great video conference backdrop) and I am perfectly capable of buying and programming ESP8266's or ESP32's.... But it's a PITA. Doing it totally DIY is a lot of tinkering with a lot of different things (power, the vagueries of LED strips you get from Aliexpress, etc). I'd totally be down for buying like 5 of these units to test with, provided they weren't unreasonably more expensive than what I'd expect to pay for an ESP8266 and related gear.
- detaro 7y ago
  I've heard good things about those (but no firsthand experience myself!): https://www.tindie.com/products/electromage/electromage-pixelblaze-v2-wifi-led-controller/ https://www.tindie.com/products/electromage/electromage-pixe...
  - linsomniac 7y ago
    I've played with the Pixelblaze and it's pretty sweet. As far as going from nothing to being able to control a string of LED lights with a series of pre-programmed effects, it is really easy. That said, I ended up abandoning the project I was going to use them for: permanently installed Christmas lights that doubled as just normal outside lighting when it wasn't a holiday. I ended up, just this weekend in fact, installing just a white LED strip light for that. Which I just did on a whim yesterday afternoon, after someone was prowling up the street looking for open cars that morning.
    - ShakataGaNai 7y ago
      Amusing you say that. At one point I went through the effort of running an entire set of RGBW LED strips, in aluminum channel, along the gutter of my mothers house (I was living in an appt in the city). So that we could have some nice programmable lights for xmas and other holidays. Man, that was so much more work than I expected... and while it worked at first water or other things kept futzing up the system (even with heavy doses of sealant near all electrical connections). Haven't tried turning the system on in more than a year now.
theamk 7y ago
This seems like way overcomplicated design - why do they need two CPUs? They claim "real time control", but this doesn't check out: their main CPU, esp8266, has a DMA that can emit stable protocol timings. And the higher level effects don't care about millisecond or ten of jitter. I am running two such simple strips right now, and it works very well.
- leggomylibro 7y ago
  ESP8266s aren't very pleasant to work with, and in my experience it can be hard to accomplish much in program logic while keeping the WiFi stable. But the cottage industry around these addressable LED driver boards is getting pretty ridiculous. The PixelBlaze already does a great job of controlling them from an ESP8266. And the LEDs aren't that hard to use with generic hardware like Arduinos, even for people with zero prior experience. There's no doubt about it, this world caters to the hucksters and the unrealistically optimistic.
  - jeffhuys 7y ago
    Regarding "in my experience it can be hard to accomplish much in program logic while keeping the WiFi stable". I've had the same; sorting a big list of numbers or doing any action that takes a few seconds can make the WiFi unstable, disconnect, or not respond. I'm curious though, there are some dual-core designs with ESP8266 compatibility out there, would that help?
    - fit2rule 7y ago
      I have no problem using coroutines to keep Wifi working fluidly on ESP8266. This just requires attention, its not too difficult to keep the ESP8266 pumping data.
  - fit2rule 7y ago
    Genuinely curious what you find unpleasant about the ESP8266 to work with?
    - 7y ago
person_of_color 7y ago
Why is there a 300 LED limit?
- readmodifywrite 7y ago
  Memory and the available CPU time to run the virtual machine. There's also a fader engine to do timed fades per pixel, that eats a fair amount of RAM as well (but it's a nice to have). It was designed for smaller, but highly dynamic, projects. I figured large scale installations were already well served in the DMX/pixel mapping space.
moepstar 7y ago
Anyone looking into this (seems you can't get your hands on the hardware just now, skimming the comments?) may want to look int WLED... All you need is a NanoMCU / ESP8266, a power source and LED-strips of you choice... There even are iOS (and i think Android) apps to control the strips via Wifi, options to sync N strips etc. etc... https://github.com/Aircoookie/WLED https://github.com/Aircoookie/WLED
- moooo99 7y ago
  +1 for WLED. They have an Android app that works quite nicely. Apparently WLED also implements ways to sync to systems like Philips Hue and offers alternative interfaces like MQTT. The community is also great and will help you out where they can.
gonesilent 7y ago
This one needs some love www.heroicrobotics.com PixelPusher
fit2rule 7y ago
I worked with Wizard23 to produce something similar, called the MagicShifter: http://magicshifter.net/ http://magicshifter.net/ Its an open source, portable battery powered pocketable light synthesiser with multiple built-in programs for doing a variety of different things - from Persistence of Vision effects (very popular), to accelerometer-based level meters, vibration visualisers, and disco/strobe effects. I added MIDI to it - meaning the whole thing can be sync'ed over rtpMIDI (MIDI over Wifi is awesome and supported out of the box by most operating systems), and I also built an Arpeggiator for it too - meaning you can have your dancers wired up with shifters on their bodies, an Arpeggiator on stage with you on your synth stacks, and MagicShifters all over the place synchronised to your drum machine. ;) We (Phillip, the hardware inventor, and myself the software guy) also designed the latest-gen MS3000 PCB to be easily useful for longer LED strip experiments, and included both 5v and 3.3v onboard, with a UART breakout, for extra hacking purposes - wire it up to a console and use it as a build indicator, or just use plain old Wifi too. We have a light strip with 200 LED's wired up to a single MagicShifter and used it to create a wall clock. Pins for wiring up longer strips are exposed on one end of the board, and you can solder the strips directly without requiring further driver hardware - just change the constants in the MS3KOS to tell it how many LED's you're driving .. EDIT: Have a look at the PCB here: https://magicshifter.github.io/magicshifter.net/img/new/ms3000-pcb-and-assembled.jpg https://magicshifter.github.io/magicshifter.net/img/new/ms30... You can see the UART breakout as well as the 4 contact points for longer LED strips... It includes a REST-based interface for changing the LED effects modes, as well as presenting a web-based Pixel Editor to create your own POV images. Sources here (100% open source) for anyone who might glean some use out of our project: https://github.com/magicshifter https://github.com/magicshifter Some other nice demo videos: https://www.youtube.com/watch?v=-xoYjA06XkQ https://www.youtube.com/watch?v=-xoYjA06XkQ https://www.youtube.com/watch?v=syP8_lccIQA https://www.youtube.com/watch?v=syP8_lccIQA https://www.youtube.com/watch?v=Shac_DN2ZrU https://www.youtube.com/watch?v=Shac_DN2ZrU https://www.youtube.com/watch?v=rcGrRuRuhYU https://www.youtube.com/watch?v=rcGrRuRuhYU EDIT: you can get an MS3000 from http://hackerspaceshop.com http://hackerspaceshop.com ..
dmos62 7y ago
A custom virtual machine and wifi. As a user, I'd much more appreciate exercises of modularity than featurefullness. Not that I don't sympathize with the maker. Over-engineering can be fun.
- readmodifywrite 7y ago
  Over-engineering is indeed fun - that was part of the point of the project, it was (and remains) a labor of love. Modularity was actually a design consideration I really struggled with, but at the end of the day, I realized I didn't want to do just another Arduino clone where you put the whole thing together yourself. I designed Chromatron under the assumption that you'd eventually end up with a lot of them, in which case the Wifi and live coding are really useful features - I wanted and all-in-one solution. But for most people who only need 1 or 2 projects, it's overkill. I think that's part of the reason it didn't sell that well.
  - dmos62 7y ago
    Easy modularity can be hard, so can't blame you for going for an AIO.

80 points by luu 7y ago

I have two of these. Way faster than an Arduino if you just want to set up some dynamic lighting, and still plenty of room for tinkering.

lepouet 7y ago
where did you buy it ?
- jkestner 7y ago
  The Kickstarter. https://www.kickstarter.com/projects/1286098094/chromatron-wifi-pixel-controller https://www.kickstarter.com/projects/1286098094/chromatron-w...

paranoidrobot 7y ago

Did I just miss it, or is there no information about price, where to get it, whether it's available now, etc? It looks interesting, and if it were the right price I might actually want one. As far as I can tell there's only the signup for more info (no, thanks) - or some links to github and some social media.

readmodifywrite 7y ago
Hi, project creator here. Sadly, the poster was a little behind on catching this... The Kickstarter boards shipped over 2 years ago, and as the hardware didn't really sell well after that, it's mostly just a personal project that this point. That being said, I am actively working on a port for some stock ESP8266 boards (Adafruit Feather, NodeMCU, etc). I don't really want to sell hardware, but I'd love to see someone else using the software.

tootie 7y ago

I love that it has a terminal block. Secure connection and no soldering.

zeva 7y ago

seems like this may no longer be in production. https://kickstarter.com/projects/1286098094/chromatron-wifi-pixel-controller https://kickstarter.com/projects/1286098094/chromatron-wifi-... https://www.tindie.com/products/readmodifywrite/chromatron-wifi-led-pixel-controller/ https://www.tindie.com/products/readmodifywrite/chromatron-w...

ShakataGaNai 7y ago
Awww, that's very unfortunate. Found the repo ( https://github.com/sapphireos/chromatron https://github.com/sapphireos/chromatron ) since it's open source and there are no commits in the last 2 years.
- readmodifywrite 7y ago
  Hi! I'm the project creator. While it's true the Kickstarter boards have all sold out, the project is still alive and well. Check out the dev and esp8266 branches, there's commits as of today ;-) My long term goal for this year is to get a port running on some stock ESP8266 boards (such as NodeMCU and the Adafruit Feather Huzzah). My original hardware design was too complex to build in a cost effective way, but I'd really like to get the software in the hands of those who are interested.

ShakataGaNai 7y ago

Very awesome demo. I want to do more DIY LED lighting in my office (because it makes for a great video conference backdrop) and I am perfectly capable of buying and programming ESP8266's or ESP32's.... But it's a PITA. Doing it totally DIY is a lot of tinkering with a lot of different things (power, the vagueries of LED strips you get from Aliexpress, etc). I'd totally be down for buying like 5 of these units to test with, provided they weren't unreasonably more expensive than what I'd expect to pay for an ESP8266 and related gear.

detaro 7y ago
I've heard good things about those (but no firsthand experience myself!): https://www.tindie.com/products/electromage/electromage-pixelblaze-v2-wifi-led-controller/ https://www.tindie.com/products/electromage/electromage-pixe...
- linsomniac 7y ago
  I've played with the Pixelblaze and it's pretty sweet. As far as going from nothing to being able to control a string of LED lights with a series of pre-programmed effects, it is really easy. That said, I ended up abandoning the project I was going to use them for: permanently installed Christmas lights that doubled as just normal outside lighting when it wasn't a holiday. I ended up, just this weekend in fact, installing just a white LED strip light for that. Which I just did on a whim yesterday afternoon, after someone was prowling up the street looking for open cars that morning.
  - ShakataGaNai 7y ago
    Amusing you say that. At one point I went through the effort of running an entire set of RGBW LED strips, in aluminum channel, along the gutter of my mothers house (I was living in an appt in the city). So that we could have some nice programmable lights for xmas and other holidays. Man, that was so much more work than I expected... and while it worked at first water or other things kept futzing up the system (even with heavy doses of sealant near all electrical connections). Haven't tried turning the system on in more than a year now.

theamk 7y ago

This seems like way overcomplicated design - why do they need two CPUs? They claim "real time control", but this doesn't check out: their main CPU, esp8266, has a DMA that can emit stable protocol timings. And the higher level effects don't care about millisecond or ten of jitter. I am running two such simple strips right now, and it works very well.

leggomylibro 7y ago
ESP8266s aren't very pleasant to work with, and in my experience it can be hard to accomplish much in program logic while keeping the WiFi stable. But the cottage industry around these addressable LED driver boards is getting pretty ridiculous. The PixelBlaze already does a great job of controlling them from an ESP8266. And the LEDs aren't that hard to use with generic hardware like Arduinos, even for people with zero prior experience. There's no doubt about it, this world caters to the hucksters and the unrealistically optimistic.
- jeffhuys 7y ago
  Regarding "in my experience it can be hard to accomplish much in program logic while keeping the WiFi stable". I've had the same; sorting a big list of numbers or doing any action that takes a few seconds can make the WiFi unstable, disconnect, or not respond. I'm curious though, there are some dual-core designs with ESP8266 compatibility out there, would that help?
  - fit2rule 7y ago
    I have no problem using coroutines to keep Wifi working fluidly on ESP8266. This just requires attention, its not too difficult to keep the ESP8266 pumping data.
- fit2rule 7y ago
  Genuinely curious what you find unpleasant about the ESP8266 to work with?
  - 7y ago

person_of_color 7y ago

Why is there a 300 LED limit?

readmodifywrite 7y ago
Memory and the available CPU time to run the virtual machine. There's also a fader engine to do timed fades per pixel, that eats a fair amount of RAM as well (but it's a nice to have). It was designed for smaller, but highly dynamic, projects. I figured large scale installations were already well served in the DMX/pixel mapping space.

moepstar 7y ago

Anyone looking into this (seems you can't get your hands on the hardware just now, skimming the comments?) may want to look int WLED... All you need is a NanoMCU / ESP8266, a power source and LED-strips of you choice... There even are iOS (and i think Android) apps to control the strips via Wifi, options to sync N strips etc. etc... https://github.com/Aircoookie/WLED https://github.com/Aircoookie/WLED

moooo99 7y ago
+1 for WLED. They have an Android app that works quite nicely. Apparently WLED also implements ways to sync to systems like Philips Hue and offers alternative interfaces like MQTT. The community is also great and will help you out where they can.

gonesilent 7y ago

This one needs some love www.heroicrobotics.com PixelPusher

fit2rule 7y ago

I worked with Wizard23 to produce something similar, called the MagicShifter: http://magicshifter.net/ http://magicshifter.net/ Its an open source, portable battery powered pocketable light synthesiser with multiple built-in programs for doing a variety of different things - from Persistence of Vision effects (very popular), to accelerometer-based level meters, vibration visualisers, and disco/strobe effects. I added MIDI to it - meaning the whole thing can be sync'ed over rtpMIDI (MIDI over Wifi is awesome and supported out of the box by most operating systems), and I also built an Arpeggiator for it too - meaning you can have your dancers wired up with shifters on their bodies, an Arpeggiator on stage with you on your synth stacks, and MagicShifters all over the place synchronised to your drum machine. ;) We (Phillip, the hardware inventor, and myself the software guy) also designed the latest-gen MS3000 PCB to be easily useful for longer LED strip experiments, and included both 5v and 3.3v onboard, with a UART breakout, for extra hacking purposes - wire it up to a console and use it as a build indicator, or just use plain old Wifi too. We have a light strip with 200 LED's wired up to a single MagicShifter and used it to create a wall clock. Pins for wiring up longer strips are exposed on one end of the board, and you can solder the strips directly without requiring further driver hardware - just change the constants in the MS3KOS to tell it how many LED's you're driving .. EDIT: Have a look at the PCB here: https://magicshifter.github.io/magicshifter.net/img/new/ms3000-pcb-and-assembled.jpg https://magicshifter.github.io/magicshifter.net/img/new/ms30... You can see the UART breakout as well as the 4 contact points for longer LED strips... It includes a REST-based interface for changing the LED effects modes, as well as presenting a web-based Pixel Editor to create your own POV images. Sources here (100% open source) for anyone who might glean some use out of our project: https://github.com/magicshifter https://github.com/magicshifter Some other nice demo videos: https://www.youtube.com/watch?v=-xoYjA06XkQ https://www.youtube.com/watch?v=-xoYjA06XkQ https://www.youtube.com/watch?v=syP8_lccIQA https://www.youtube.com/watch?v=syP8_lccIQA https://www.youtube.com/watch?v=Shac_DN2ZrU https://www.youtube.com/watch?v=Shac_DN2ZrU https://www.youtube.com/watch?v=rcGrRuRuhYU https://www.youtube.com/watch?v=rcGrRuRuhYU EDIT: you can get an MS3000 from http://hackerspaceshop.com http://hackerspaceshop.com ..

dmos62 7y ago

A custom virtual machine and wifi. As a user, I'd much more appreciate exercises of modularity than featurefullness. Not that I don't sympathize with the maker. Over-engineering can be fun.

readmodifywrite 7y ago
Over-engineering is indeed fun - that was part of the point of the project, it was (and remains) a labor of love. Modularity was actually a design consideration I really struggled with, but at the end of the day, I realized I didn't want to do just another Arduino clone where you put the whole thing together yourself. I designed Chromatron under the assumption that you'd eventually end up with a lot of them, in which case the Wifi and live coding are really useful features - I wanted and all-in-one solution. But for most people who only need 1 or 2 projects, it's overkill. I think that's part of the reason it didn't sell that well.
- dmos62 7y ago
  Easy modularity can be hard, so can't blame you for going for an AIO.